Functions similar to linreg3 and hl.map in HAIL 0.2

Help [0.1]
1

Hi,

I wonder if there are functions in HAIL 0.2 that are similar to linreg3 and hl.map in HAIL 0.1.

I would like to implement the following in HAIL 0.2 -

ht = ht.annotate(‘set_of_codes = [{}].toSet()’.format(df[0]))
for x in df[1:]:
ht = ht.annotate(‘set_of_codes = set_of_codes.add({})’.format(x))
ht = ht.annotate(‘set_of_codes = set_of_codes.remove("")’)

Thanks,
Emily

1 Like
2

yes, this is going to be way easier in 0.2. What is df, though?

3

also, hl.linear_regression has all the functionality that linreg3 used to, plus a fair bit more.

4

df is actually a list.

5

should I loop through the list of phenotypes and run hl.linear_regression if I want to do a phewas?

6

Here’s the docs for linear regression in 0.2: https://hail.is/docs/devel/methods/stats.html#hail.methods.linear_regression

One difference between linreg3 and linreg in 0.1 is that linreg3 was capable of doing multiple phenotypes at once. In 0.2 this is possible too! Just pass a list of phenotype expressions as the y parameter.

7

I’m also having a bit of trouble understanding what you’re trying to do with the code in your original post – another example might help.